Metal core doping refers to the precise introduction of heterogeneous metal atoms into a metal core framework, achieving direct regulation of the cluster's electronic structure, geometric configuration, and stability. Precisely controlling the number of doped atoms is the primary prerequisite for the precise synthesis of metal nanoclusters; it directly determines the cluster's electronic structure and geometric configuration, thereby influencing its intrinsic optoelectronic properties.^[43-46]. In the process of metal core doping, the type of doped atom significantly affects the quantity of doping. Taking the icosahedral Au₂₅(SR)₁₈cluster (Au = gold, SR = thiolate ligand) as an example, when doping with Pd (palladium), Pt (platinum), Cd (cadmium), or Hg (mercury), the cluster tends to incorporate only a single heteroatom.^[47]. The team of Zhikun Wu^[48]introduced single Pd and Pt atoms into Au₂₅(PET)₁₈(PET = SCH₂CH₂Ph), successfully synthesizing Au₂₄Pd(PET)₁₈and Au₂₄Pt(PET)₁₈clusters. Spectral analysis indicated that single-atom doping significantly altered the cluster's electronic structure, subsequently changing the magnetism of the cluster from paramagnetic before doping to diamagnetic after doping. The team of Zhongning Chen^[49]prepared Ag₆Cu clusters by doping Cu(I) (Cu = copper) into Ag₇(Ag = silver) clusters; their photoluminescence quantum yields (PLQY) increased from 27% for Ag₇to 91% for Ag₆Cu (Figure 1a). This is primarily due to the enhanced structural rigidity of the cluster after Cu doping, as well as accelerated rates of intersystem crossing (ISC) and reverse intersystem crossing (RISC). Circularly polarized organic light-emitting diodes (LEDs) fabricated based on this material achieved an external quantum efficiency (EQE) of up to 26.7%, superior to most red organic LEDs. Doping with a single heteroatom has certain limitations in regulating the electronic and geometric structures of clusters; increasing the number of doped atoms can achieve more significant regulation of the cluster's geometric structure. The team of Quanming Wang^[50]doped six Cu atoms into Au₂₂, increasing the cluster's PLQY from 9% to nearly 100%; this is because after doping with six Cu atoms, the cluster structure contracted, the highest occupied molecular orbital-lowest unoccupied molecular orbital (HOMO-LUMO) energy gap decreased, and the ISC rate accelerated (Figures 1b, c). This work provides important material and methodological support for the development of ultra-high brightness and high-efficiency single-cluster electroluminescent devices.

Precise control of doping sites is another key dimension for revealing the structure-activity relationships of clusters and achieving targeted performance regulation. The Maran team^[51]combined nuclear magnetic resonance spectroscopy and electrochemical analysis to precisely resolve the site selectivity of different dopant atoms in Au₂₅: Pt and Pd prefer to occupy the central position of the icosahedral Au₁₃core, while Cd and Hg tend to dope at the icosahedral shell sites (Figure 1d). This difference in site selectivity is mainly attributed to the inherent physicochemical properties of the dopant atoms, including atomic radius and electronegativity. Professor Zhu Yan's team^[52]compared the catalytic performance of a series of Ag-doped Au clusters with different doping sites in the cycloaddition reaction of CO₂with epoxides. Experimental results show that the Au₁₉Ag₄cluster with fully exposed silver sites exhibits the best catalytic activity and stability, while the Au₂₀Ag₁cluster with partially exposed silver sites has the lowest catalytic efficiency. This is mainly because the Au₁₉Ag₄cluster possesses Ag sites with high structural openness and partial positive charge, which can effectively promote the adsorption of CO₂and the ring-opening of epoxides (Figure 1e). Such precise design and regulation of active sites are of great significance for constructing high-performance single-cluster catalytic sensors and energy conversion devices.

In summary, metal core doping enables atomically precise manipulation of the electronic structure and geometric configuration of clusters by precisely controlling the number, type, and sites of dopant atoms, laying a solid material foundation for the directional optimization of cluster optoelectronic properties and the design of high-performance single-cluster optoelectronic devices.

As the interface where ligands directly interact with the external environment for clusters, their precise modification is one of the core strategies for regulating cluster interfacial properties and optoelectronic performance^[53-55]. Precise ligand modification mainly focuses on atomic-level control over the number of ligands and substitution sites, typically achieved by regulating reaction kinetics and thermodynamics. Professor Xie Jianping's team^[56]developed a redox reaction-guided surface engineering method, achieving reversible addition/removal of single ligands on Au₂₅clusters, clarifying the reversible interconversion mechanism between [Au₂₅(MHA)₁₈]^-(MHA = 6-mercaptohexanoic acid) and [Au₂₅(MHA)₁₉]⁰. Studies indicate that after adding an extra thiol ligand, the stability of the ligand shell of Au₂₅clusters changes; this change is speculated to stem from alterations in the ligand shell packing mode and internal stress. In addition to precisely doping the original ligands in the cluster by number, the team also utilized 6-aza-2-thiothymine, L-arginine, and tetraoctylammonium as ligands to perform stepwise modification of Au₁₀clusters via self-assembly (Figure 2a), significantly improving the photoluminescence efficiency of the clusters^[57]. This research demonstrates that precise numerical control and functional modification of surface ligands on clusters can achieve specific optimization of cluster photophysical properties. In application research oriented towards devices, Professor Yao Hongbin's team^[58]precisely introduced ligands with different electron-donating abilities into copper iodide clusters, finding that as the electron-donating ability of the ligands increased (dimethylamino > pentyl > methyl), the emission wavelength redshifted and the bandgap decreased, proving that ligand electronic effects can precisely regulate emission color (Figure 2b). The PLQY of the modified material approached 90%, and the EQE of LEDs fabricated based on this material was significantly superior to most lead-free metal halide LEDs, fully demonstrating the great potential of ligand functionalization in realizing efficient and spectrally tunable single-cluster light-emitting devices.

Different ligand sites on the cluster surface possess specific chemical microenvironments, electronic coupling strengths, and spatial accessibility^[59-60]. Achieving precise modification of ligands at specific sites is of great significance for revealing the mechanism of cluster-environment (e.g., electrode) interface interactions and optimizing the performance of single-cluster devices. The Ackerson team^[61]controlled the reaction time to utilizep-BBT (p-BBT = p-bromobenzenethiol) for Au₁₀₂(p-MBA)₄₄ clusters (p-MBA = p-mercaptobenzoic acid) to perform ligand exchange. By employing a kinetic control strategy, the ligand exchange reaction was made to occur preferentially at specific sites with kinetic advantages, achieving precise control over the ligand modification sites. The specificity of doping sites can also be achieved through the physicochemical properties of the doping ligands themselves. The Teranishi team^[62]exchanged ligands between [Au₂₅(SC₂Ph)₁₈]^-(SC₂Ph = 2-phenylethanethiol) and porphyrin thiolates (Por) with different steric hindrances, finding that bulky porphyrin thiolates are more likely to exchange to positions on the cluster surface with smaller steric hindrance, namelyFigure 2c, at the top S₂ site of the cluster's staple structure shown. This is the result of the combined effects of steric repulsion and thermodynamic stability: the larger spatial openness of the S₂ site reduces steric conflicts between ligands, enabling bulky ligands to undergo ligand exchange reactions in a lower-energy configuration and remain stably on the cluster surface. Niihori et al.^[63]reacted Au₂₄Pd(SC₂Ph)₁₈ clusters with different ligands. The results indicated that during ligand exchange, the reaction occurs preferentially at thiolate ligand sites directly connected to the metal core. This is mainly due to the strong electronic structure coupling between the ligands at these sites and the metal core, small steric hindrance, and low reaction energy barriers. Therefore, introducing functional ligands at highly reactive sites can directly and efficiently regulate the electron density of the metal active centers and optimize the carrier mobility at the cluster-electrode interface.

In summary, the precision of ligand functionalization is primarily reflected in the accurate control over the number and sites of modification. By regulating the number and sites of ligand modification, it is possible to enhance the intrinsic optoelectronic properties of clusters, which is of decisive significance for improving the charge transport efficiency, interface stability, and optoelectronic catalytic performance of single-cluster devices.

It should be noted that although the metal core doping and ligand functionalization strategies discussed in this section have mostly been validated in cluster aggregate systems, the regulation of intrinsic properties such as the electronic structure, energy level alignment, and interface stability of individual clusters constitutes the physical basis determining the performance of single-cluster devices. Therefore, the aforementioned precise synthesis strategies provide a critical material foundation for the subsequent performance optimization of single-cluster devices.

One of the core challenges in precisely fabricating single-cluster devices is constructing electrode gaps to accurately match cluster dimensions. Since most cluster sizes fall within the <2 nm range, the precision of traditional lithography techniques is insufficient to meet these requirements.^[66]. Electromigration technology is currently one of the most commonly used techniques for fabricating static single-cluster devices.^[67], its core principle utilizes Joule heating and electromigration forces to jointly induce directional migration of atoms within metal nanowires, ultimately resulting in a controlled fracture at a preset location to form electrode gaps on the sub-nanometer to few-nanometer scale (Figure 3a)^[68].

Subsequently, the solution containing the target clusters is drop-cast or spin-coated onto prepared nanogap electrodes (mostly gold electrodes). By leveraging terminal functional groups on the cluster surface ligands, such as thiol (—SH) and phosphino (—PR₃), covalent bonds are formed with the gold electrodes, allowing individual clusters to spontaneously and selectively bridge the two ends of the nanogap, forming a stable "electrode-cluster-electrode" junction^[71]. To improve the precision and success rate of single-cluster bridging, a self-assembled monolayer of molecules or atoms can be pre-modified or deposited in situ on the electrode surface. Single-cluster junctions precisely constructed based on this technology can be further integrated with gate electrodes to form a single-cluster field-effect transistor (FET) structure^[72-73], thereby achieving in situ and precise regulation of the cluster's charge state and transport behavior, greatly expanding its applications in fields such as logic operations and sensing.

Single-cluster devices fabricated via electromigration technology are static devices. Their significant advantages include highly stable electrode structures, facilitating precise measurements under extreme conditions such as low temperatures. Furthermore, the fabricated single-cluster devices exhibit good reproducibility and are easy to integrate. Therefore, this remains one of the most commonly used precise fabrication techniques for current research on static single-cluster devices. However, this method also has drawbacks, including reliance on complex micro-nano fabrication processes (resulting in higher costs), a relatively low success rate for single-cluster bridging, and difficulty in precisely controlling the electrode gap size.

The precise construction and regulation of the cluster-electrode interface are core factors determining the electron transport properties of single-cluster junctions. Among these, the type of anchoring element at the cluster-electrode interface plays a decisive role in the electrical transport process. Taking chalcogen elements, which are most widely used in clusters, as an example: as elements that can coordinate with both metal elements within the cluster and the working electrode, the type of chalcogen element significantly affects the electrical transport behavior of single-cluster junctions. Professor Nuckolls' research group^[78]synthesized a series of isostructural cobalt chalcogenide clusters: (1) Co₆Te₈(P(C₂H₅)₃)₆ (Co = cobalt, Te = tellurium); (2) Co₆Se₈(P(C₂H₅)₃)₆ (Se = selenium); (3) Co₆S₈(P(C₂H₅)₃)₆. STM-BJ results show that conductance follows the order Te > Se > S (Figure 4a). Cyclic voltammetry (CV) characterization demonstrates that Te-anchored clusters possess the lowest oxidation potential and the highest HOMO energy level. Combined with Density Functional Theory (DFT) calculations, it was found that the HOMO is concentrated on Te atoms. According to the Hard and Soft Acids and Bases (HSAB) theory, among the three elements, the softest base Te achieves optimal orbital overlap with Au, effectively reducing the interfacial barrier and thereby resulting in higher cluster conductance; conversely, the hardest base S leads to the lowest cluster conductance value. In addition to chalcogen elements, halogens as anchoring elements can also directly regulate the electrical transport behavior of single-cluster junctions. The team of Hong Wenjing^[85]synthesized AgCuX clusters coordinated by different halogens (X = Cl (chlorine), Br (bromine), I (iodine)) and compared the influence of different halogens on the electrical transport behavior of single-cluster junctions. Halides possess lone pair electrons; the electrostatic potentials (ESP) around them exhibit negative potential, and their Average Local Ionization Energy (ALIE) values are low. This allows them to form stable connections with unsaturated sites on Au electrodes through electrostatic attraction and chemical bonding (Figure 4b). Raman spectroscopy confirmed the existence of Au—X bonds. ESP-ALIE analysis further indicated that halogen atoms and their surrounding regions are the most active anchoring areas. Thus, the authors determined that halogens can serve as anchoring sites to form stable connections with gold electrodes. Moreover, the coupling strength sequence with gold electrodes is Au—Cl < Au—Br < Au—I. The conductance sequence of the three clusters is consistent with the coupling strength, namely AgCuCl (3.88 nS) < AgCuBr (8.30 nS) < AgCuI (14.43 nS). This is because from Cl to I, the atomic radius increases and electronegativity decreases, improving orbital overlap with Au, enhancing interfacial coupling, and increasing electron transmission probability. Furthermore, in this work, through rational structural design utilizing the exposed lone pair electrons of halogens, the authors achieved direct bonding between the electrode and the cluster interface. This avoided the system complexity caused by multi-site anchoring within the cluster, resulting in a more concentrated conductance distribution and significantly improving the clarity and reliability of the device structure. At the cluster-electrode interface, besides the type of element at the anchoring site, the steric effect of ligands is also an important factor regulating the electrical transport behavior of the cluster-electrode interface. Xavier et al.^[86]used Co₆Se₈(PEt₃)₆clusters as a model, introducing ligands containing sulfur-methyl groups (L2~L4) into them. The phosphorus end of the ligand coordinates with cobalt, while the sulfur-methyl end coordinates with the gold electrode. Ligands L2 and L4 adopt a para-substituted configuration (Figure 4c). DFT calculations revealed that para-substituted ligands (L2, L4) can form effective π-conjugated electron pathways connecting the cluster core with the anchoring group, thereby causing clusters 2 and 4 to exhibit distinct conductance peaks (Figure 4d). In contrast, ligand L3 is meta-substituted; in diethyl-3-(thiomethyl)phenylphosphine, the 3-(thiomethyl)phenyl group cannot rotate around the P—C bond to form an effective conductive pathway, resulting in no significant conductance signal for cluster 3. Comparing clusters 1~3 reveals that the ligand substitution position can affect electrical transport efficiency by controlling the spatial orientation of the electron transmission path. The Nuckolls research group^[78]also compared Co₆Te₈(P(n-Bu)₃)₆(P(n-Bu)₃ = tri-n-butylphosphine) and Co₆Te₈(P(C₂H₅)₃)₆ , two types of Co₆Te₈clusters with ligands of different steric hindrance, finding that as the steric hindrance of the ligands increases, the cluster conductance decreases significantly. This is attributed to the fact that bulky ligands hinder the orbital overlap between the lone pair electrons of Te and the gold electrode, thereby weakening the interfacial coupling strength.

In addition, the solvent environment and ligand length^[87-90]and other factors can also regulate the cluster-electrode interface coupling and electrical transport behavior to varying degrees. In summary, by precisely designing the types of anchoring elements at the cluster-electrode interface (to regulate coupling strength) and the spatial configuration/steric hindrance of ligands (to regulate the geometric structure and electron transport pathways at the interface), researchers can achieve directional and predictable regulation of the electrical transport behavior of single-cluster junctions. Specifically, "directional" refers to the systematic adjustment of parameters such as conductance and coupling strength of single-cluster junctions through rational design of the anchoring element types and ligand spatial configurations at the cluster-electrode interface; "predictable" refers to predicting the electrical transport behavior of clusters with different structures based on existing experimental results combined with theoretical simulations, providing a solid theoretical foundation for designing high-performance single-cluster optoelectronic devices.

The intrinsic structure of clusters (such as size and atomic composition) is the intrinsic determining factor of their electrical transport behavior. In the synthesis and design of clusters, although macroscopic studies have shown that structural changes significantly affect properties such as catalysis and luminescence of clusters^[91-94], the average effects at the macroscopic scale and size heterogeneity seriously hinder the precise construction of their structure-activity relationships^[95]. However, the construction of single-cluster devices provides an ideal platform for revealing the regulation of intrinsic structural features, such as size effects and atomic doping, on the electrical transport behavior of clusters at the atomic level.

The size of the metal core affects its quantum confinement effect and electronic density of states, thereby enabling the regulation of its electrical transport capability. Gunasekaran et al.^[96]synthesized model clusters with a single Co core and Co core dimers. Comparison revealed that increasing the size of the cluster metal core can significantly enhance the device current and lead to current saturation phenomena (Figure 5a). The authors pointed out that the increase in metal core size reduces the reorganization energy of the system and leads to a denser distribution of energy levels, synergistically enhancing the electron tunneling probability and transport efficiency, resulting in current enhancement and saturation behavior. Subsequently, the Hong Wenjing group compared Ag clusters with continuously varying sizes^[97](Ag_x, wherex = 25, 43, 44, 63, 78, 136, 141, and 374) regarding their electrical transport behavior. They found that the size effect of the metal core in single-cluster junctions exhibits a trend completely opposite to that in organic single-molecule junctions; specifically, conductance increases monotonically with increasing cluster size (Figure 5b), with a conductance decay constant of approximately 0.4 nm^-1. Experiments and theoretical calculations indicate that the HOMO-LUMO gap of Ag_xclusters decreases significantly with increasing metal core size, and the coupling between the cluster and electrodes is enhanced, leading to increased conductance. On the other hand, within the same cluster, conductance also exhibits an increasing trend with the increase in the linear distance between electrode anchoring sites. This study demonstrates that by precisely controlling the cluster size, the gap magnitude and interface coupling strength can be effectively regulated, thereby achieving directional modulation of its electrical transport behavior.

Introducing heteroatoms into clusters for atomic-level precise doping is a powerful means to regulate their electronic structures. Yuan et al.^[98]utilized STM-BJ technology to investigate the influence of central atom doping on the electrical transport behavior of Anderson-type polyoxometalate (POM) clusters. They synthesized four types of POM clusters with different central metal atoms: MoFe, MoCo, MoNi, and MoZn. Experimental results demonstrated that by changing the central metal atom, the conductance of POM clusters could be directionally regulated within an order of magnitude (Figure 5c). Furthermore, by introducing a temperature gradient in the STM-BJ, the authors measured the thermoelectric potential of the clusters and obtained their Seebeck coefficients. They found that the Fermi levels of the MoFe, MoNi, and MoZn clusters were close to the HOMO, while the Fermi level of the MoCo cluster was close to the LUMO. Combined with theoretical calculations, it was revealed that the substitution of the central atom led to a non-uniform reconstruction of the electrostatic potential distribution within the clusters. This electron structure change induced by atomic-level precise doping not only directly regulated the cluster conductance but also affected its bias dependence and thermoelectric response. This study directly correlated atomic-level doping, fine-tuning of electronic structures, and multi-dimensional transport properties at the single-cluster level, providing atomic-scale insights for designing functional nanoelectronic devices based on doped clusters.

In summary, the intrinsic structure of clusters is a core factor regulating the electrical transport behavior of single-cluster devices. These structure-activity relationships obtained at the single-cluster level not only deepen the understanding of fundamental laws of charge transport at the nanoscale, but also lay a solid theoretical and experimental foundation for precisely customizing single-cluster devices with specific optoelectronic properties at the atomic level, effectively promoting the application of single-cluster devices in fields such as nanoelectronics, quantum information processing, and energy conversion.

The significant Coulomb blockade effect in clusters lays the theoretical foundation for realizing electrically controlled switching devices. Based on this effect, changing the charge state of clusters through external regulation can achieve the "on" and "off" states of the device^[15]. Lovat et al.^[100]used STM-BJ technology to perform bias regulation on Co₆S₈L₆ clusters (L = diethyl-4-(methylthio)phenylphosphine). They found that charge transport was significantly suppressed within a bias range of 360 to -450 mV (Figure 6a), while outside this range, the current of the single-cluster junction increased significantly with bias. The maximum switching ratio of single-cluster switching devices designed based on this bias regulation mechanism can reach 600. The core strategy for achieving switching functionality is to regulate the charge state of clusters through external stimuli. In addition to bias regulation, electrochemical potential regulation is also an effective means to achieve switching functionality. The Vezzoli team^[101]prepared single-cluster junctions via the STM-BJ method, which exhibited obvious conductance responses under electrochemical potential control. As the potential moved from zero to negative values, the conductance increased slightly; when the potential reached approximately -0.3 V, the cluster accepted an electron, its charge state changed from -3 to -4, and the conductance decreased significantly. When the potential moved in the positive direction, the cluster charge state switched from -3 to -2, and the conductance dropped to the lowest level. This electrochemical potential regulation achieved reversible switching among three charge states: -4, -3, and -2 (Figure 6b), with the switching ratio between adjacent states exceeding one order of magnitude. Whether through bias regulation or electrochemical potential regulation, the essence of both is regulating the charge state of clusters to achieve the injection or removal of single electrons. These works fully demonstrate that utilizing single clusters holds promise for designing high-performance, multi-state molecular switches to achieve single-electron manipulation at the atomic level.

Applying single-cluster devices to the field of transistors relies on utilizing the quantum effects of clusters to achieve single-electron control. Such devices are typically fabricated using techniques like electromigration or chemical deposition combined with "soft landing." In 2017, the team led by Song Fengqi first conducted electrical transport measurements on Au₁₃clusters^[72], observing Coulomb blockade behavior in this cluster at 1.6 K.I-V(Irepresents current,Vrepresents voltage) curves show that the conductivity of Au₁₃is suppressed at low bias voltages, followed by step-like current jumps at higher voltages; the switching ratio of this transistor approaches 5 (Figure 6c). When the temperature rises to 100 K, theI-Vcurve exhibits linear behavior, indicating that Coulomb blockade disappears at temperatures above 100 K. It is worth noting that the electrical behavior of the aforementioned transistors was observed at low temperatures, which limits their application at room temperature to some extent. To overcome this limitation, the Vondel team^[102], after fabricating nanogaps using electromigration technology, deposited charged aluminum clusters into the electrode gaps via "soft landing" technology, observing the Coulomb blockade effect at room temperature. Its charging energy is 0.14 eV, far greater than the thermal energy at room temperature (approximately 26 meV). This key progress confirms the feasibility of single-cluster-based transistors operating normally at room temperature, marking an important step toward their industrialization and integration.

Designing in-situ catalytic detection devices based on single clusters can avoid the averaging effects in traditional catalytic experiments, thereby enabling a more accurate understanding of elementary catalytic reactions at the single-cluster and single-molecule levels, and providing theoretical and experimental guidance for designing efficient electrocatalysts.^[103-104].

Ma Wei Team^[84]Constructed a dynamic electrochemical device based on single Au₂₅²⁺clusters using single-particle collision electrochemistry and applied it to the oxygen reduction reaction (ORR). This technique successfully captured the dynamic structural evolution of Au₂₅²⁺during the ORR process (Figure 6d): desorption/re-adsorption of surface ligands and conformational changes of the Au₂₅⁹⁺core jointly caused the device current signal to exhibit characteristic "on-off" state switching and "on" state fluctuations. Recently, the team used the same method to study the catalytic behavior of different single-atom doped silver nanoclusters (M₁Ag₂₄, M = Ag, Au, Pt, Cu) in ORR^[105]. This platform successfully distinguished the differences in ORR activity among individual M₁Ag₂₄clusters caused by different dopant atoms (M) and their positions (central C site or surface O site): clusters doped with Au and Pt at the center exhibited the highest activity, while those doped with Cu on the surface showed lower activity, attributed to the specific regulation of the cluster electronic structure by the dopant atoms. This technology not only provides a powerful tool for atomic-level precise electrochemical research of single clusters but also demonstrates the potential of these devices as fast, high-throughput screening tools for nanocatalysts, greatly expanding the application scope of single-cluster devices.

When using a single metal nanocluster to fabricate light-emitting diodes, the spectral broadening problem caused by uneven particle distribution in aggregated materials can be effectively avoided. Furthermore, by directionally optimizing the composition, structure, or excitation conditions of the cluster, the luminous efficiency and device stability of single-cluster light-emitting diode devices can be significantly enhanced.

Dickson team^[73]Innovatively adopted pulse excitation technology to utilize silver nanoclusters (Ag₂-Ag₈) in symmetric nanojunctions to achieve single-cluster LED functionality, breaking through the limitation of non-polarity in nanodevices under traditional DC/AC excitation. The team discovered that high-frequency pulse/AC excitation can significantly improve the spectral and thermal stability of electroluminescence in silver nanoclusters, with a spectral width far narrower than that under DC excitation, effectively extending the LED lifespan. This work established a correlation between pulse parameters and electroluminescence performance at the nanoscale, providing effective experimental basis for optimizing the efficiency of nano-LED devices. Furthermore, the highly efficient response of silver nanoclusters in this device lays a solid material foundation for developing high-stability nanoscale light sources.

The optoelectronic applications of the aforementioned single-cluster devices not only highlight their significant value in bridging atomic-level structures with macroscopic device functionalities, but also lay a solid foundation for their future applications in fields such as nanoelectronics, quantum information processing, and precision catalysis.